It is generally agreed that ERa is expressed mainly in the stromal compartment, but not in the epithelial compartment of prostate. However, results from stromal-epithelial tissue recombination experiments gave distinct results as to whether epithelial ERa or stromal ERa plays a role in the estrogen-mediated responses in prostatic tissue. The experiment, in which prostatic compartments were reconstituted by tissue recombination and transplanted under the kidney capsule, concluded that both epithelial and stromal ERa are required for estrogen-mediated squamous metaplasia; yet, the field debated that the stromal or epithelial cells isolated from the anterior prostate, which has a different developmental origin from ventral prostate and dorsalateral prostate of animals. In addition to prostate development, it has been reported that early exposure of E2 will lead to imprinting effects of mouse prostates to develop squamous metaplasia (SQM). Furthermore, a growing body of evidence indicated that estrogen plus androgen will stimulate the carcinogenesis of prostate in recombinant experiments and in intact animals. Interestingly, ERa, but not ER, is critical for E2-imprinting squamous metaplasia and carcinogenesis using conventional ERaKO and ERKO mice as model or material source to conduct experiments. The conventional total ERa knockout (ERaKO) mice may generate compound effects and sometimes it is unclear whether the observed phenotypes on a given tissue is due to the ERa function in the given tissue or due to the defect of ERa function in other tissues. Therefore, it is important to establish a better material source, or a better animal model to re-examine the role of stromal and epithelial ERa in prostatic homeostasis and carcinogenesis. We decided to generate the tissue specific KO mice by using the Cre-LoxP strategy to elucidate the respective roles of epithelial and stromal ERa in prostatic homeostasis and carcinogenesis. The main goal of this proposal is to obtain a better understanding of the roles of ERa in prostatic biology, and to identify the potential ERa target genes in the prostatic tissue. To achieve these goals, we will pursue the following Specific Aims. Aim 1 will use Cre-LoxP strategy to generate mice with epithelial-ERaKO and to determine whether epithelial ERaKO will affect prostate development and estrogen imprinting effect. Aim 2 will generate mice with prostate stromal-ERa knockout and investigate prostate development and the estrogen imprinting effects. Aim 3 will establish mouse prostate cancer models with ERa knockout and investigate prostate tumorigenesis and tumor progression. Aim 4 will identify and examine how ERa regulates the expression of the stromal factors and target genes that may mediate the estrogenic effects on aberrant prostate epithelial growth, squamous metaplasia and cancerous transformation.